Quick Start

Get your first AI agent running in 5 minutes.

Prerequisites

• An API key from any supported LLM provider (OpenAI, Anthropic, Google, xAI, or a local model)
• macOS, Linux, or Windows (WSL)

Install

Script (recommended):

curl -fsSL https://get.droid.plus | sh

Docker:

docker pull sinapticai/droid:latest

Binary download:

Download the binary for your platform from GitHub Releases.

Platform	Architecture	Download
Linux	amd64	droid-linux-amd64
Linux	arm64	droid-linux-arm64
macOS	Apple Silicon	droid-darwin-arm64
macOS	Intel	droid-darwin-amd64
Windows	amd64	droid-windows-amd64.exe

Create Your First Agent

Option A: AI Builder UI (no code)

Start Sinaptic® DROID+ and open the visual builder:

droid up
# Open http://localhost:8081/builder/ in your browser

The AI Builder UI lets you configure your agent visually — pick a model, write a system prompt, add tools, and test it live. When you're done, it generates the YAML config for you.

Option B: YAML Config

Initialize a new project:

droid init my-agent
cd my-agent

This creates a project directory with:

my-agent/
├── droid.yaml          # Runtime config (ports, LLM keys, security)
├── configs/agents/
│   └── my-agent.yaml   # Your agent definition
├── .env.example        # API key template
└── data/               # Runtime data (RAG, logs)

Edit .env.example → .env with your API key:

cp .env.example .env
# Edit .env and add your OPENAI_API_KEY (or any other provider key)

Start the runtime:

droid up

You'll see:

 ____  ____   ___  ___ ____
|  _ \|  _ \ / _ \|_ _|  _ \  _
| | | | |_) | | | || || | | |(_)_
| |_| |  _ <| |_| || || |_| |  _|
|____/|_| \_\\___/|___|____/  (_)

Sinaptic.AI DROID+ v0.5.0
Agent runtime engine starting...
✓ Loaded 1 agent: my-agent
✓ SinapticAI security: enabled (community mode)
✓ OpenAI-compatible API: http://localhost:8080/v1
✓ Management API: http://localhost:8081
✓ Agent Builder UI: http://localhost:8081/builder/

Talk to Your Agent

Sinaptic® DROID+ exposes an OpenAI-compatible API, so you can use any OpenAI SDK or just curl:

curl:

curl http://localhost:8080/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "my-agent",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Python (OpenAI SDK):

from openai import OpenAI

client = OpenAI(base_url="http://localhost:8080/v1", api_key="any")

response = client.chat.completions.create(
    model="my-agent",
    messages=[{"role": "user", "content": "Hello!"}]
)

print(response.choices[0].message.content)

Node.js (OpenAI SDK):

import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'http://localhost:8080/v1',
  apiKey: 'any',
});

const response = await client.chat.completions.create({
  model: 'my-agent',
  messages: [{ role: 'user', content: 'Hello!' }],
});

console.log(response.choices[0].message.content);

The model field maps to the agent name in your YAML config. The api_key can be any string for local development (unless you've configured authentication in droid.yaml).

Agent Config Explained

Here's a minimal agent config (configs/agents/my-agent.yaml):

name: "my-agent"
description: "Customer support assistant"

model:
  name: "gpt-4o-mini"       # Any model from your configured provider
  max_tokens: 1024
  temperature: 0.7

personality: |
  You are a helpful customer support assistant.
  Be concise and accurate. Always be polite.

tools:
  - name: "current_time"
    type: "builtin"

sinaptic:
  enabled: true              # Enable security checks

Key fields:

• name — unique agent identifier, used as the model in API calls
• model.name — the LLM model to use (e.g., gpt-4o-mini, claude-sonnet-4-20250514, gemini-2.0-flash)
• personality — system prompt that defines the agent's behavior
• tools — built-in tools, REST API endpoints, or MCP servers the agent can use
• sinaptic.enabled — toggle SinapticAI security (prompt injection detection, PII masking)

CLI Commands

droid up          # Start the runtime (loads all agents)
droid down        # Gracefully stop a running instance
droid init NAME   # Create a new agent project
droid agents      # List loaded agents and their status
droid health      # Check server health
droid version     # Print version, commit, and build date

Next Steps

• LLM Providers — Configure OpenAI, Anthropic, Gemini, Grok, or local models
• MCP Tools — Add external tools via Model Context Protocol
• SinapticAI Security — Understand the built-in security layer
• Docker Deployment — Run Sinaptic® DROID+ in production with Docker
• Editions — Compare Community, Pro, and Enterprise features